Interregional Migration of the Poor: Some Recent Changes

c3. Wee PAO-73

|NTERREGIONAL

MIGRATION OF

THE POOR:

Some Recent

Changes

arry H. Long

CURRENT POPULATION

REPORTS

Special Studies

Series P-23, No. 73

U.S. DEPARTMENT

OF COMMERCE

Bureau

■INTERREGIONALMIGRATION OFTHE POOR:

Some RecentChanges

by

Larry H. Long

Issued November 1978

/jl\U .S. Department of CommerceJuanita M. Kreps, SecretaryCourtenay M. Slater, Chief Economist

BUREAU OF THE CENSUSManuel D. Plotkin, Director

BUREAU OF THE CENSUS

Manuel D. Plotkin, Director

Robert L. Hagan, Deputy Director

Daniel B. Levine, Associate Director for

Demographic Fields

POPULATION DIVISION

Meyer Zitter, Chief

Library of Congress Cataloging in Publication Data

Long, Larry H

Interregional migration of the poor.

(Current population reports : Special studies :

Series P-23; no. 73)

Supt. of Docs. no.: C 3.186:23/73

1. Migration, Internal—United States. 2. Poor—

United States. 3. Public welfare—United States.

4. Southern States—Economic conditions. l. Title.

II. Series: United States. Bureau of the Census.

Current population reports : Special studies : Series

P-23; no. 73.

HA203.A218 no. 73 [HB1965] 312'.0973s 78-606112

[301.32’6'0973]

For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C.

20402. Postage stamps not acceptable; currency submitted at sender's risk. Remittances from foreign

countries must be by international money order or by draft on a U.S. bank. Current Population

Reports are sold in two subscription packages: Series P-20, P-23, P-27, and P-60 are available for

$40.00 per year ($10 additional for foreign mailing); Series P-25, P-26, and P.28 are available for

$70.00 per year ($17.50 additional for foreign mailing). The single-copy price of this report is $1.40.

PREFACE

This is another in a series of analytical studies undertaken by demographers

in the Population Division, Bureau of the Census. A distinguishing feature of

these occasional publications is that they include broad speculative analysis and

illustrative hypotheses by the authors as an aid in understanding the statistics

and assessing their potential impact on public policy. The usual scope of these

studies is probably broader than that of annual census reports on population

subjects but less complete than book-length monographs.

Previous publications in the analytical series include: Some Recent Changes

in American Families, by Paul C. Glick (1975); The Geographical Mobility of

Americans: An International Comparison, by Larry H. Long and Celia G.

Boertlein (1976); and Racial Succession in Individual Housing Units, by Larry

H. Long and Daphne Spain (1978). Additional studies are in preparation.

The author received his Ph.D. degree in 1969 from the University of Texas

at Austin, where he was affiliated with the Population Research Center. He

spent 1969-70 on a postdoctoral fellowship at the Population Studies Center,

University of Pennsylvania. Since August 1970, Dr. Long has been with the

Census Bureau, where much of his work has involved regional analysis and

internal migration. Future work is planned to develop more fully themes of

spatial aspects of welfare and poverty.

Work on the present study began several years ago and was initially carried

out by Lynne R. Heltman, now of the U.S. Commission on Civil Rights.

Computer programming was done by Norman Kaplan and Emmett Spiers.

The research reported in this study was done in part with funds from the

Center for Population Research, National Institute of Child Health and Human

Development.

INTERREGIONAL MIGRATION OF THE POOR:

SOME RECENT CHANGES

Does the South still have net outmigration of persons below the poverty

level? This question has concerned policy makers and others for quite some

time, and it is a question that will persist. At issue is the degree to which

poverty is and has been interregionally transferred through migration. This

study offers an answer to this question and explores some of the implications

of that answer.

Public debate on the question became particularly intense during the 1960's,

when migration from the South was often cited as an important cause of the

rising number of welfare recipients in northern States (see references in Long,

1974; and Long and Heltman, 1975). The debate continues, since persistent

high levels of welfare dependence in the North are interpreted as partly

reflecting the lingering effects of past South-to-North migration (see, for

example, the discussions presented in Time, 1975; Business Week, 1976; U. S.

News and World Report, 1976; Lyons, 1976; and Business Week, 1977).

Futhermore, even as the South shifted to net inmigration in the 1960's, there

was evidence of continuing high levels of outmigration among persons with

limited education and skills. Hence, it is entirely possible that as the South

shifted to overall net in migration, it continued to have net outmigration of

persons below the poverty level. Partly because of this potential for poverty

redistribution through migration, welfare policies are increasingly looked upon

as a matter for national concern rather than a strictly regional or local issue.

To address these matters, policy makers need data of many types, but a

crucial piece of information involves knowing the poverty status of migrants at

approximately the time of migrating. Such information would allow one to

assess whether, on an annual basis, a region was having net inmigration or net

outmigration of low-income persons who May be likely to put at least a

temporary burden on welfare agencies in places of destination. Data of this

type have not been available from past censuses or other data sources. Recent

censuses have asked about birthplace and residence 5 years before the census

date, and while data from these questions have been extremely valuable in

identifying migration streams, they have not allowed one to assess the income

level of migrants at the time of moving. Since the census question asked about

residence 5 years earlier and income in the calendar year preceding the census,

the data reflect the income position of migrants up to 5 years after migration.

A few past surveys collected data on income and migration over nearly the

same interval of time, but the data were never tabulated to show income of

1

migrants in the regional streams. Creating the data needed for this study

entailed going back to surveys taken in the late 1960's, retabulating them to

show poverty status of migrants around the time of migrating, and cumulating

the results over a period of years in order to increase the effective sample size.

The results were then compared with more recent surveys, which are larger and

have more detailed questions on residential background. The purpose was to

develop data that would allow one to decompose net and gross migration

streams into poverty and non-poverty components based on each migrant's

income very close to the time of the migration.

Such data are useful not only for the immediate purpose of answering

whether on a current basis a given region has net inmigration or net

outmigration of the poor, but also for evaluating various theories regarding

motivations behind the migration of low-income populations. Interstate

differences in size of welfare payments are frequently cited (see references in

Long, 1974) as having an important effect on the migration of the low-income

population, the theory being that low-income persons are motivated to move

from areas where payments are low (notably the southern States) to areas

where payments are high (notably the northern States). If this theory has

validity in accounting for migration of the poor, then one probably expects to

find a rather steady net outmigration of the poor from the South, for there has

been little narrowing of North-South differences in average size of welfare

benefits. The often-cited differences between Mississippi and New York State

in the size of AFDC payments (Aid to Families with Dependent Children) has

actually widened over the last few years; in 1970 the average AFDC family in

New York received over five times what the average AFDC family in Mississippi

received, but by 1976 the average New York AFDC family was receiving nearly

eight times the amount received by the average Mississippi AFDC family

(National Center for Social Statistics, 1970, table 7; 1976, table 4). In view of

the persistence of large benefit differences between the South and the North,

failure to find net outmigration from the South of low-income persons in the

1970's would certainly call into question the current validity of the theory that

interstate differences in size of welfare payments are acting as an inducement

for the low-income population to migrate.

Another use of the data developed for this study is in evaluating the effects

of interregional migration on family and per capita income in various regions of

the country. Reports from the Census Bureau (1976a, p. 6) and the Bureau of

Economic Analysis (see Bretzfelder, 1977) have shown the southern region to

be experiencing stronger and more consistent increases in personal income in

the 1970's than the other major regions. Showing whether recent southern

income gains result from net outmigration of low-income persons would

contribute to understanding the processes underlying the regional income

changes.

A Brief Background

The South's total net migration—irrespective of migrants’ income level—is

illustrated in figure 1 for periods between 1880 and 1975. In this figure and in

all the tables that follow, the South is defined according to census practice and

includes States as far north as Delaware and as far West as Texas. A map of the

four major regions is included as figure 2.

Throughout much of the last century and well into this century, large parts

of the South exhibited characteristics commonly associated with depressed

areas. Even after World War II and into the 1950's, much of the region could

still be described as essentially agrarian, and a population that was poor,

equipped with inadequate educations, and faced with declining employment

opportunities. The understandable outcome of such circumstances over a long

period of time was substantial migration from the South to other regions of the

country.

But what is perhaps surprising about southern migration is that the exodus

did not begin earlier. In spite of poor economic conditions, the region had very

little net outmigration before the 1910-20 decade, and most of the net

outmigration during that decade probably occurred as a consequence of the

North’s industrial development in connection with World War I. The most

likely explanation for the small volume of migration from the region before

World War I is that the expanding number of jobs in northern industries went

disproportionately to immigrants. Thus, as long as immigrants were pouring

into the North from the 1880’s to around the time of World War I, there was

little incentive for surplus labor in the South to move North, only to have to

compete for jobs with immigrants. Instead, potential migrants opted to stay in

the South, where family and friends were located and could provide mutual

(though meager) support.

In the 1920's migration from the South increased. This was a decade when

immigration was sharply reduced and industrial jobs in the North were fairly

plentiful. But this level of migration from the South was not sustained, for the

drying up of economic opportunities in the North during the depression of the

1930's was associated with a decline in southern outmigration.

This brief background to post-World War II southern migration was provided

simply to emphasize that at the time of mobilization for the war, the South

had a large reservoir of rural, mostly agricultural population whose movement

north had in previous decades been restrained first by the presence of large

numbers of immigrants and then by the depression of the 1930's. The

mitigation of these two migration-inhibiting factors and the rapid increase in

3

FIGURE 1.

The South's Net Migration from 1880 to 1975

Net Migration (millions)

3.0

2.5

2.0

1.5

1.0

0.5

2.624

.740

—.236

–0.5

—1.0 —.985

—1.091

—1.540

–2.0

–2.241

–2.5

–3.0

—1.4 16

1910

1920

1880–

1890

1890

1900

1900

1910

1920

1930

1930

1940

1940

1950

Source:

1950

1960

1960

1970

1970

1975

Data for decades up to 1940-1950 are from Eldridge and Thomas (1964, p. 41). Data for

1950-1960 are from Bowles and Tarver (1965, p. 2). Data for 1960-1970 are from Bowles

et al. (1977, p. 4). Data for 1970-1975 are from U.S. Bureau of the Census (1976b, p. 19).

Note:

Prior to 1950-1960, data on net migration are for persons 10 years old and over. More

recent data include persons under 10.

industrial jobs associated with World War II was accompained by a large

volume of migration from the South. During the 1940-50 decade the South

had a net outmigration of 2.2 million persons.

It is interesting to speculate that had not large numbers of immigrants

arrived in the North to take industrial jobs and had the depression of the

1930's not been so severe or so prolonged, the outmigration of surplus

agricultural labor from the South might have been a smoother, more gradual

process and more easily handled by the northern cities to which southern

migrants primarily went. Instead, migration from the South was rather abrupt,

reaching very high levels during and after World War II and continuing into the

1950’s.

By the 1960's, however, many parts of the industrial North were

experiencing a relative slowdown in economic activity. Many jobs were going

South as a number of industries relocated to southern States where taxes were

often lower and where land and labor were generally cheaper. Also, improved

transportation in the South (usually in the form of more and better highways)

encouraged the relocation of employers wishing to modernize or update their

manufacturing plants. As a result, the South shifted to net inmigration in the

1960's—the first sizable net inmigration experienced by the region in nearly a

century. The shift to net inmigration in the 1960's was the product of more

southerners staying in the region of their birth, more former outmigrants

returning, and more northerners and westerners going to the South (see Long

and Hansen, 1975; 1977a). The South's gain from migration has increased

dramatically in the 1970's; in the 5 years from 1970 to 1975 the South's net

inmigration was over three times the size of its net inmigration in the entire

1960-70 decade.

Even as the South shifted to net inmigration, there was evidence that the

region was continuing to have substantial outmigration of persons with low

levels of education (Fein, 1965; Bowman and Myers, 1967), leaving open the

possiblity that in the 1960's the South was having overall net inmigration (see

figure 1) but having net outmigration of persons below the poverty level. If this

was the case, then the region was continuing to contribute, on an annual basis,

to poverty levels elsewhere.

Recent Migration

In order to see if the South's change to overall net inmigration had also

brought net inmigration of the poor, several surveys carried out in the 1960's

were examined, beginning with the Survey of Economic Opportunity, taken in

1967 by the Census Bureau for the Office of Economic Opportunity and for

the specific purpose of measuring poverty and some of its concomitants. The

measure of poverty used by the Federal government for statistical purposes was

first developed in 1965 (Orshansky, 1965) and takes into account family size,

age and sex of family head, and whether residence is farm or nonfarm in

determining if total annual money income is above or below a predetermined

poverty level. The poverty-nonpoverty cutoff points are usually published in

statistical reports on poverty (e.g., U. S. Bureau of the Census, 1973, p. 457;

1977, p. 198). In order to take inflation into account, the poverty cutoff levels

are annually adjusted according to changes in the Consumer Price Index. In

1970 (i.e., 1969 income) the poverty cutoff for a nonfarm family of four

persons was $3,748; in 1977 (income in 1976) the poverty threshold for such a

family was $5,815.

The 1967 Survey of Economic Opportunity included several questions on

residential background, one of which was the respondent's place of birth.

These data were tabulated to show persons 14 years old and over who were

born in the South but living in other regions in 1967; these persons were

lifetime outmigrants whose poverty status (based on total money income

received in 1966) could be contrasted with that of lifetime inmigrants (persons

14 and over born outside the South but living in the South in 1967). These

data were accepted as an indicator of the degree to which southern migration

had contributed to poverty levels in other regions in 1967.

Useful as such data May be, they do not show interregional poverty flows on

an annual basis, and in order to estimate these flows, special tabulations were

made from Current Population Surveys taken in March of 1968, 1969, 1970,

and 1971. Each of these surveys had questions on the respondent's place of

residence in the preceding March and total money income received in the

calendar year preceding each of the March surveys. Thus, income and migration

were being measured over closely overlapping intervals of time, thereby

providing a measure of poverty status of migrants around the time of

migrating.

Data from the four Current Population Surveys were cumulated in order to

show for the four regions for the 1967-71 period average annual inmigration,

outmigration, and net migration according to migrants’ poverty status. More

recent data of comparable nature were obtained from tabulations of Current

Population Surveys taken in March 1976 and March 1977; each survey had

questions on place of residence in March 1975 and total money income

received during the calendar year preceding the survey. The results from the

several surveys are shown for the South in table 1.

As of 1967, the South had clearly sent more low-income persons to other

regions than it received in return. According to data from the 1967 Survey of

Economic Opportunity, there were 941,000 low-income Southern-born

7

TABLE 1. The South's Change to Net Inmigration of

Low-Income Persons Between 1967 and 1977

(Numbers in thousands)

All

Migration flows income "*" Poorpoor

levels

Migrants to South

(1) Living in South in 1967 but born in

other regions (persons 14 years and

over) 4,570 4,103 467

(2) 1967-71, annual average, persons

1 year old and over 1,263 1,076 187

(3) 1975-76, persons 1 year old and

over 1,192 1,014 178

(4) 1975-77, persons 2 years old and

over 1,932 1,621 311

Migrants from South

(5) Born in South but living in other

regions in 1967 (persons 14 years

and over) 7,485 6,544 941

(6) 1967–71, annual average, persons

1 year old and over 1,255 1,025 231

(7) 1975-76, persons 1 year old and

over 990 889 100

(8) 1975-77, persons 2 years old and

over 1,470 1,287 184

Net migration

Up to 1967 = (1) - (5) –2,915 –2,441 –474

1967-71 = (2) - (6) 8 51 -44

1975-76 = (3) - (7) 202 125 78

1975-77 = (4) - (8) 462 334 127

NOTE: Poverty status is computed from total money income re

ceived in the calendar year preceding each of the March surveys.

SOURCE: Data shown on lines (1) and (5) are from Bowles et al.

(1973, pp. 112-117). Data on lines (2), (3), (6), and (7) are from March

Current Population Surveys, which asked for respondents' residence

one year earlier. Data on lines (4) and (8) are from the March 1977

CPS, which asked about residence in March 1975.

persons 14 years old and older living below the poverty line in other regions of

the country. Since the total number of low income persons 14 and over living

outside the South was 9,005,000, one can conclude that in 1967 the South

had contributed at least 10 percent of all persons in poverty outside the South.

Having sent 941,000 low-income persons 14 and over to other regions, the

South received only 467,000 low-income inmigrants in return, for an apparent

net outmigration of 474,000 low-income persons. Thus, the massive migration

from the South up to 1967 had definitely contributed to poverty in other

regions.

By 1967, the South had probably shifted to overall net inmigration, that is,

net inmigration irrespective of income levels of migrants. But as the region

shifted to overall net inmigration, it seems to have continued for a few years to

have net outmigration of persons below the poverty level. The continuation in

1967-71 of net outmigration of persons below the poverty level is indicated in

the set of figures in the lower right-hand corner of table 1, which shows that

for the 1967-71 period the region had an annual average net outmigration of

44,000 persons below the poverty level.

Sometime between then and 1976, however, the region experienced an

unrecognized shift to net inmigration of persons below the poverty level. Look

again at the figures in the lower right-hand corner of table 1. For a 12-month

period in 1975-76 the region is shown to have had net inmigration of 78,000

persons below the poverty level, and this indicator of a change in trend is

confirmed by the figures for 1975-77, when the region is estimated to have had

net inmigration of 127,000 persons below the poverty level. These figures

highlight a significant shift, whereby the South, a low-income region

traditionally exporting low-income persons to other parts of the country, came

to have a small net gain of persons below the poverty level.

Changes in net migration are usually the result of changes in the numbers of

inmigrants as well as changes in the numbers of outmigrants. In the case of the

South's shift to net inmigration of the poor, the change seems to result more

strongly from declines in the number of poor outmigrants rather than increases

in the number of poor inmigrants. For example, the number of low-income

inmigrants to the South hardly changed at all between 1967-71 and 1975-76,

but the number of low-income outmigrants appears to have declined from an

annual average of 231,000 in 1967-71 to 100,000 in 1975-76. Thus, the change

seems to come from the region's retaining more of its own poor rather than

drawing off the poor of other regions. Such a conclusion would be consistent

with the earlier finding that the South as a whole shifted from net

outmigration in 1955-60 to net inmigration 1965-70 more as a result of greater

retention of Southern-born persons than greater inmigration of persons born in

other regions (see Long and Hansen, 1975).

But what about the other regions? In particular, have the northeastern

States, long thought to be strongly attracting the poor of other regions, also

experienced a change in net migration of the poor? Gross and net migration for

the three nonsouthern regions are given in table 2 for various periods between

1967 and 1977.

The data suggest that by the mid-1970's the Northeast had net outmigration

of persons below the poverty level. Back in 1967-71 the number of low-income

inmigrants to the Northeast approximately equaled the number of low-income

outmigrants, although table 2 shows a net outmigration of the poor of about

16,000 per year—a number too small to be significant. But for a 1-year period

in 1975-76 the region's net outmigration of the poor is shown to be 40,000,

and for a 2-year period in 1975-77 net outmigration of the poor is shown to be

171,000. Rather clearly, the region is now sending more low-income persons to

other regions than it is receiving in return—an important change.

The North Central region probably had net outmigration of the poor

continuously between 1967 and 1977. Each of the three measurement

periods–1967-71, 1975-76, and 1975-77—indicated that the North Central

region was having net outmigration of persons below the poverty level. There is

no clear evidence of change in this regard.

The West is shown to have had net inmigration of the poor between 1967

and 1977. For each of the measurement intervals, the number of low-income

migrants to the West appears to be greater than the number of low-income

outmigrants, giving the region net inmigration of the poor, Thus, of the four

major regions, only the West consistently had net inmigration of the poor

between 1967 and 1977.

One might summarize these statistics by saying that insofar as interregional

migration of the poor in the mid-1970's is concerned, the North Central region

and the West seem to be continuing a pattern in existence in the late 1960's.

The North Central region seems to be continuing to have net outmigration of

persons below the poverty level, and the West seems still to be having net

inmigration of the poor. The data do not provide a basis for detecting change

in the net migration of the poor for the North Central or West regions. For the

South and the Northeast, however, there is some evidence of change. The

South is shown to have shifted to net inmigration of the poor by 1977, and the

Northeast seems to have come to experience net outmigration of the poor by

1977.

10

TABLE 2. Gross and Net Migration According to Migrants'

Poverty Status, for the Three Nonsouthern Regions:

Various Periods from 1967 to 1977


All

Interregional migration income Non- POOR

poorlevels

NORTHEAST

Inmigrants

1967–71, annual average, persons 1 year

old and over 558 495 64

1975-76, persons 1 year old and over 454 416 38

1975-77, persons 2 years old and over 669 604 64

Outmigrants




1975-77, persons 2 years old and over 1,283 1,048 235

Net migration


old and over -177 -160 –16

1975-76, persons 1 year old and over –212 -173 -40

1975-77, persons 2 years old and over –614 –444 -171

NORTH CENTRAL

Inmigrants





Outmigrants


old and over 1,028 864. 163

1975-76, persons 1 year old and over 1,023 843 181


11

TABLE 2. Gross and Net Migration According to Migrants'

Poverty Status, for the Three Nonsouthern Regions:

Various Periods from 1967 to 1977–Continued


- - All Non

Interregional migration Income Poor

poorlevels

NORTH CENTRAL–Continued

Net migration


old and over -116 –68 -48

1975-76, persons 1 year old and over –355 –256 -100

1975-77, persons 2 years old and over –362 –310 -52

WEST

Inmigrants


old and over 1,014 818 196



Outmigrants





Net migration





NOTE: Poverty status is computed from total money income re

ceived in the calendar year preceding each of the March surveys.

SOURCE: Current Population Surveys taken in March 1968, 1969,

1970, 1971, 1976, and 1977.

Doublechecking the Results

It is plausible to suppose that some of the above migration patterns could be

influenced by the movement of retirees, some of whom May have pensions so

meager as to classify the recipients below the poverty level. Since retirees on

small pensions May be motivated to move to the South, where living costs are

frequently quite low, it is possible that low-income retirees account for the

South's recent net inmigration of persons below the poverty level. In order to

test for this possibility and simply to verify the results from tables 1 and 2,

tabulations were prepared from the 1976 Survey of Income and Education

(SIE).

For the present study, the SIE has many advantages, one of which is its

large size. In approximately 151,000 households, each individual was asked

how long he or she had lived in the State of usual residence at the time of the

interview and the State of previous residence for those whose duration of

residence was less than a lifetime. For persons reporting a duration of residence

of less than 1 year, tabulations were prepared in order to duplicate as closely as

possible the 1-year migration data shown in tables 1 and 2, except to separately

identify the migration of persons in the labor force ages. That is, the attempt

was to use an independent data source, the SIE, to verify the results from the

1976 and 1977 Current Population Surveys and at the same time focus on the

migration of persons in the labor force ages. For persons 20 to 64 years of age,

table 3 shows inmigrants and outmigrants for each of the four regions during

the 1 year preceding the 1976 SIE, which was conducted from April through

July and based poverty status on income received during the calendar year

1975.

The results from table 3 serve to validate previous conclusions and

demonstrate that they apply as well to persons in the labor force ages. The

table shows that in the 12 months preceding the 1976 SIE, the Northeast and

North Central regions had net outmigration of low-income persons at ages 20

to 64, and the South and West regions had net inmigration of low-income

persons 20 to 64 years old. The significance of these conclusions is not simply

to confirm earlier results, but to demonstrate that low-income retirees are not

the “explanation” of the South's change to net inmigration of the poor. The

South's net inmigration of the poor in the mid-1970's is the result of

movements of persons at economically active ages.

Another consideration that merits mentioning is the possible role of return

migration. It is, of course, possible that the South has started to pull back

many of the low-income migrants who left in earlier decades. There is,

unfortunately, no way of directly testing for this effect. Had the SIE produced

13

TABLE 3. Low-Income Persons Among Each Region's

Inmigrants and Outmigrants in the 12 Months Preceding

the 1976 Survey of Income and Education, for Persons

20 to 64 Years Old


All income levels Below poverty level

Region Inmi- Outmi- Net Inmi- Outmi- Netrants grants migra- grants grants migra

9 tion tion

South 922 658 264 113 75 38

West 676 446 230 87 46 41

Northeast 281 566 –285 22 67 –45

North Central 462 672 –210 59 92 –33

data on State of birth, one could have assessed the role of return migration

simply by disaggregating the southern inmigrants shown in table 3 according to

those born in the South (returnees) and others. Although the SIE included a

question on country of birth, if failed to obtain the State-of-birth data that

would have allowed the identification of return migrants.

There are, however, two reasons for downplaying the role of return

migration in the South's shift to net inmigration of the poor. First, as

mentioned in an earlier section, there has been no increase in the number of

low-income southern inmigrants (returnees and others); the change in

the South's net migration of the poor seems to have resulted largely from fewer

poor outmigrants. Had returnees played a major role in the change in net

migration, one might have expected to find increased inmigration of poor

persons to the South, but this did not happen between 1967-71 and 1975-76.

A second reason for doubting that return migration has played a major role in

shifting the South to net inmigration of the poor is that in the past return

migrants generally have not been those with the lowest levels of education. One

study used 1970 census data to define a return migrant as a person born in the

South and living in the South in 1970 but not in 1965; with these data, the

study found that southern outmigrants who were most likely to go back to the

South were those with the highest level of education, especially college

graduates (see Long and Hansen, 1977b). As a result of this positive

14

educational selectivity in return movement, the returnees actually had higher

levels of education than the initial outmigrants, who in turn had higher levels

of education than persons who had stayed in the South. This relatively high

level of education among returnees suggests persons who could most readily

take advantage of expandingjob opportunities and not persons likely to add to

the poverty rolls.

Duration of Residence

A way of extending the above analysis and obtaining a fuller evaluation of the

effects of migration on regional poverty levels is to assess the poverty status of

migrants at successive periods after migrating. For example, a crucial variable

May be a region's ability to absorb large numbers of low-income migrants (even

net inmigration of the poor) but provide employment opportunities that

enable the poor to become nonpoor after a fairly short period of residence.

Because it collected data on duration of residence, the SIE provides a basis

for assessing the poverty status of migrants not only around the time of

migrating, but also after having lived for varying periods of time in the region

of destination. For each of the four regions, this information is shown in table

4 for interregional migrants 20 to 64 years old. Also shown for comparison is

the poverty status of (1) each region’s total population 20 to 64 years old, (2)

lifetime nonmigrants (persons 20 to 64 who had always lived in the same

State), and (3) persons 20 to 64 who had moved between States within the

same region.

The idea here is to try to see to what extent low-income migrants to

different regions of the country become nonpoor with increasing length of

residence. Look first, however, at the percent poor in the different migration

streams. It May come as a surprise to persons who have looked upon the

Northeast as getting more than its share of low-income migrants to note that

migrants going to the Northeast in the 12 months preceding the survey seem to

be somewhat less likely to be poor than migrants going to the other three

regions. About 7.8 percent of working-age persons moving to the Northeast in

the 12 months preceding the 1976 SIE were below the poverty level; among

migrants going to the other three regions, the percent poor was higher—

between 12 and 13 percent. A similar pattern seems also to have existed in

1967-71 (see tables 1 and 2). Whatever the reason, migrants to the Northeast

between 1967 and 1976 were generally somewhat less likely than migrants to

the other three regions to be poor at the time of migrating.

15

o,

TABLE

4.Poverty

Status

ofNonmigrants

and

Migrants

Who

Have

Lived

inEach

Region

for

Varying

(Numbers

inthousands)

Periods

ofTime,

for

Persons

20

to64

Years

Old:

1976

Total

Moved

Migrants

toeach

region

;.

Always

between

-

ation

lived

States

-

Moved

Moved

Moved

Moved

Regions

20

to64

insame

but

in

Mººn

1-3

4-5

6-9

10

years

State

Same

months

years

years

years

ormore

old

region

ago

ago

ago

years

ago

SOUTH

All

income

levels

37,149

19,913

8,384

922

1,150

734

1,115

3,149

Below

poverty

4,090

2,612

724

113

8360

76

233

Percent

below

poverty

11.0

13.1

8.6

12.3

7.2

8.2

6.8

7.4

WEST

All

income

levels

21,161

6,575

4,553

676

719

555

839

5,118

Below

poverty

1,723

529

340

87

60

46

64

338

Percent

below

poverty

8.1

8.0

7.5

12.9

8.3

8.3

7.6

6.6

NORTHEAST

All

income

levels

27,257

16,730

4,240

281

290

280

431

2,250

Below

poverty

1,795

1,010

149

22

1715

36242

Percent

below

poverty

6.6

6.0

3.5

7.8

5.9

5.4

8.4

10.8

NORTH

CENTRAL

All

income

levels

30,894

18,671

4,808

462

487

371

617

4,231

Below

poverty

2,118

1,163

283

594834

51368

Percent

poverty

6.9

6.2

5.9

12.8

9.9

9.2

8.3

8.7

NOTE:

Included

inthe

total

but

not

shown

separately

are

persons

whose

last

residence

was

outside

the

United

States

and

persons

who

failed

toprovide

complete

information

on

previous

residence.

The

latter

group

constituted

0.8

percent

ofpersons

20–64

years

old.

SOURCE:

1976

Survey

ofIncome

and

Education.

For every region—including the Northeast—very recent inmigrants (those

who have lived in the region for less than 1 year) are more likely to be poor

than the total destination population. In other words, every region tends to

attract migrants who are more likely to be poor than is “average” for the

region (compare columns 1 and 4 in table 4). Equally important is that

migrants who have lived in a region for a few years are less likely to be poor

than the very recent arrivals (compare column 4 with columns 5, 6, and 7).

This seeming improvement after a few years in the economic status of

low-income inmigrants results not only from income gains experienced by

those who stay, but also from some low-income migrants simply moving on to

other regions. Importantly, however, in each of the four regions, migrants of 1

to 10 years’ residence are less likely to be poor than more recent arrivals,

suggesting that each region during this time was able to reduce poverty among

migrants who moved in and stayed.

There is little evidence to suggest, though, that poverty level of migrants

steadily goes down with increasing length of residence. Columns 4 through 8 of

table 4 represent migrants with increasing duration of residence in each of the

regions, and the data in these columns do not support the idea that the percent

poor progressively goes down with longer durations of residence. Instead, there

is an erratic pattern. The data May mean that many low-income inmigrants

either manage to become nonpoor after a few years or else move on to other

regions; those who do not fit these descriptions but stay for more than a few

years perhaps experience episodes of poverty, alternating between work and

welfare. A true test of such a scenario requires longitudinal data which are not

available.

Note especially the migrants with the longest duration of residence. The last

column of table 4 shows persons who have lived in the same region for 10 years

and whose previous residence was in a different region. For these migrants,

their last interregional move occurred in 1966 or earlier, and they presumably

have had a long while in which to adjust and accommodate to the region of

destination. In the South, the West, and the North Central regions, the

migrants with 10 years' residence are less likely to be poor than the very recent

arrivals (less than 1 year's residence). In the Northeast, however, the migrants

of 10 years’ residence seem to have a greater incidence of poverty than the

more recent arrivals. This group of long-term migrants in the Northeast May,

indeed, be the “residue” of the pre-1967 migration from the South. The

relatively high incidence of poverty among pre-1967 migrants to the Northeast

May be evidence to support the idea that the region still does carry a poverty

burden (perhaps even a welfare burden) from an earlier migration. But the

more recent migration pattern is to reduce this burden, for since 1967 the

Northeast has not had net inmigration of the poor from other regions, and in

17

the last few years the Northeast has been a net exporter of low-income persons

to other parts of the United States.

The poverty status of persons who have never moved between States is given

in column 2 of table 4, and it is interesting to note that in the North as in the

South these lifetime nonmigrants constitute the majority of persons below the

poverty line. In the Northeast, 56 percent of the low-income population 20 to

64 years old consists of persons who have never moved between States.

Similarly, 55 percent of the North Central region’s low-income population of

working age are lifetime nonmigrants. Of the four regions, the South has the

largest proportion of “home-grown” poverty–64 percent of its low-income

population 20 to 64 years old has never moved between States. Only in the

West did nonmigrants constitute a minority of the low-income population. In

the West 31 percent of both the poor and nonpoor populations had never

moved between States. These data are useful in demonstrating that the

“typical” poor person in the North, like the “typical” poor person in the

South, is an individual who has never made an interstate move.

Effects on Regional Income

A use of the data developed in earlier sections is to evaluate sources of

income growth in the various regions of the country. A number of Census

Bureau reports using data from the annual Current Population Survey have

pointed to recent faster-than-average rates of income growth in the South and

mildly speculated that this was due to outmigration of the southern poor. As

one Bureau report commented (U.S. Bureau of the Census, 1976a, p. 6):

Income changes in the four regions (Northeast, North Central, South,

and West) reflect endogenous changes as well as interregional migration.

Distilling the separate influences of the two sets of factors, however, is

difficult. For instance, a strong outmigration of low income southerners

could have the effect of raising median income in the South and lowering

the median in non-South regions.

It could, but the data in tables 1, 2, and 3 suggest rejecting this implicit

hypothesis and accepting the alternative hypothesis, namely, that faster

than-average income growth in the South results from income-generating forces

within the region (the “endogenous changes” mentioned above).

Perhaps it is useful to illustrate the faster-than-average rate of income

growth in the South and to speculate about when North-South gaps in income

and poverty might close if present trends continue, particularly in view of

recent changes in regional net migration of the poor. For these purposes, table

5 was assembled to show per capita income and poverty levels in the four

18

TABLE 5. Per Capita Income and Poverty Level, by Region,

for Selected Years: 1940 to 1977

Region

and Year

Average annual

Per capita Poverty percentchange

income level since preced

(dollars) ing date

Abso- As Number Per- Per Number

percent of poor ... of poorlute cent capita

level - persons poor income persons

national (000's) (000's)

average

UNITED STATES

1940

1950

1960

1970

1977

SOUTH

1940

1950

1960

1970

1977

WEST

592 100.0 (NA) (NA) (NA) (NA)

1,496 100.0 (NA) (NA) 9.3 (NA)

2,215 100.0 38,685 22.1 3.9 (NA)

3,920 100.0 26,931 13.6 5.7 —3.6

6,399 100.0 24,975 11.8 8.2 -1.1

389 65.7 (NA) (NA) (NA) (NA)

1,138 76.1 (NA) (NA) 10.7 (NA)

1,747 78.9 19,105 35.6 4.3 (NA)

3,362 85.8 12,326 20.2 6.6 –4.4

5,776 90.3 10,354 15.2 9.0 -2.5

700 118.2 (NA) (NA) (NA) (NA)1940

1950

1960

1970

1977

1,700 113.6 (NA) (NA) 8.9 (NA)

2,484 112.1 4,378 16.1 3.8 (NA)

4,142 105.7 3,956 11.7 5.1 -1.0

6,789 106.1 4,015 10.5 8.2 0.2

19

TABLE 5. Per Capita Income and Poverty Level, by Region,

for Selected Years: 1940 to 1977–Continued

Average annual

Per capita Poverty percent change

income level since preced

(dollars) ing date

Region As Number - Number

and Year *percent of poor Per Per . poor

lute cent capita

level ional persons poor income persons

national (000's) (000's)

average

NORTHEAST

1940 774 130.7 (NA) (NA) (NA) (NA)

1950 1,716 114.7 (NA) (NA) 8.0 (NA)

1960 2,539 114.6 6,249 14.4 3.9 (NA)

1970 4,422 112.8 4,722 9.9 5.5 –2.8

1977 6,824 106.6 4,949 10.2 7.2 O.7

NORTH CENTRAL

1940 603 101.9 (NA) (NA) (NA) (NA)

1950 1,590 106.3 (NA) (NA) 9.7 (NA)

1960 2,289 103.3 8,953 17.7 3.6 (NA)

1970 3,968 101.2 5,927 10.7 5.5 –4.1

1977 6,518 101.9 5,657 9.9 8.3 –0.7

NA. Not available.

NOTE: Poverty status is based on censuses or surveys taken in

March or April of each of the years shown and is computed from total

money income received during the preceding calendar year. Per

capita income refers to the calendar year shown, except that per

capita income for 1977 actually refers to 1976.

SOURCE: Column 1 is from Bureau of Economic Analysis (1973,

pp. 232-241; and Bretzfelder, 1977, p. 17). Column 3 is from unpub

lished tabulations from the 1960 Census and U.S. Bureau of the

Census (1973, table 1; and 1977, table 19).

20

regions from 1940—when southern outmigration increased greatly—up to 1977.

The data tend to justify looking upon the South as the Nation's traditional

depressed area. It is interesting to note that of the four major regions, the

South is the only one to have lower-than-average per capita income. The other

three regions have had higher incomes than the national average at least since

1940.

In fact, until recently southern incomes were much below the national

average. In 1940 per capita income in the South was slightly less than

two-thirds the national average, and at that date the average southerner's

income was only one-half that of the average resident of the Northeast. In view

of such large regional income differentials in 1940, the incentive to leave the

South was understandably great, and in the decade following 1940 the South

experienced the largest net outmigration in its history (refer back to figure 1).

The massive outmigration in that decade was associated with rapid growth in

southern income, which rose from 66 percent of the national average in 1940

to 76 percent in 1950.

Southern outmigration remained fairly high in the 1950's, but growth in per

capita income in that decade was relatively modest, rising from 76 percent of

the national average in 1950 to 79 percent in 1960. The 1960's represented a

turnaround decade for the South insofar as migration is concerned, for the

region came to have net inmigration for the first time in nearly a century. But

the South's net inmigration in the 1960's was net inmigration of nonpoor

persons, for the region continued to have net outmovement of persons who

were poor at approximately the time of migration. Partly because of these

countervailing net movements of the poor and nonpoor populations in the

1960's, the South made important income gains, its per capita income rising

from 79 percent of the national average in 1960 to 86 percent in 1970.

In the 1970's the relative income gap between the South and the non-South

is closing more rapidly than in the 1960’s. By 1976 per capita income in the

South had risen to 90 percent of the national average, and if this rate of change

is maintained, by 1980 per capita income in the South will be 93 percent of

the national average. The high rate of income growth in the South in the

1970's is particularly interesting because it comes at a time when the region

shifted to net inmigration of persons below the poverty level—a condition

which might ordinarily have a depressing effect on rates of increase in overall

personal incomes. From 1940 to 1970 the South’s income gains were certainly

due in part (though how large a part is unknown) to the region's having net

outmigration of persons below the poverty level.

21

These data support Garnick’s thesis that in many ways the recent shift of

personal income out of the Northeast and to southern States is an acceleration

of processes that have been underway for quite some time (see Garnick, 1978).

Earlier researchers had questioned the permanency of the shifts, however,

pointing out that the South in 1970 still had many large reservoirs of poverty

(see Jusenius and Ledebur, 1976). Although the South even in 1977 had a

higher incidence of poverty than any of the other three regions, substantial

income gains since 1970 have helped to dry up many of the 1970 reservoirs of

poverty. Consider that in 1970 the average southerner was nearly twice as

likely to be poor as the average person living outside the South; but according

to CPS data for 1977 the average southerner’s likelihood of being poor was

only around 50 percent greater than that of a person living outside the South

(these calculations are based on data compiled for table 5). Clearly, the

incidence of poverty in the South has been falling rather rapidly in the 1970’s

in spite of the region’s shift to net inmigration of the poor during this period.

Given these rapid increases in per capita income and declines in poverty in

the South in the 1970's, it is interesting to speculate as to when the

long-standing income gaps between the South and the non-South might close if

present trends were to continue. Such an extrapolation of the post-1970

pattern can be made, not as a forecast or prediction, but as an illustration of

the magnitude of recent changes. For example, from table 5 observe that in the

1970's the number of poor persons in the South has been reduced by nearly 2

million, from 12.3 million in 1970 to 10.4 million in 1977. At the same time,

the number of poor persons living outside the South hardly changed at all,

numbering about 14.6 million in both 1970 and 1977. These figures, based on

data from the March 1977 Current Population Survey, mean that just about all

of the reduction in poverty in the United States since 1970 has occurred in the

South—historically the nation’s poorest region.

If the average annual rates of change in the poor and nonpoor populations

in the South and non-South were to continue at the same rates as observed in

the 1970-77 period, the historic poverty gap between the South and the rest of

the Nation would be completely eliminated in 11 years. That is, in 1988

poverty in the South would be down to 9.4 percent of the population and

would be about the same as in the rest of the country. These calculations were

made simply by computing the average annual rates of change in the poor and

nonpoor populations of the South and non-South in 1970-77 and applying the

rates of change to the 1977 base populations. The results indicated that the

percent poor in the South and in the rest of the country would converge at 9.4

percent in 11 years, or 1988.

If the South/Non-South poverty gap were in fact to be eliminated in 1988,

22

that milestone would be reached 50 years after President Roosevelt’s 1938

declaration (quoted in Moynihan, 1978, p. 4):

It is my conviction that the South presents right now the nation's No. 1

economic problem—the nation’s problem, not merely the South's. For

we have an economic unbalance in the nation as a whole, due to this

very condition of the South's. It is an unbalance that can and must be

righted, for the sake of the South and of the Nation . . .

These remarks of President Roosevelt were offered as justification for

numerous New Deal programs designed to upgrade living standards in the

South and elsewhere through public-works projects, and Federal spending

patterns are alleged still to benefit the South (Havemann et al., 1976).

If the post-1970 patterns were to continue, not only would the South/non

South poverty gap be closed by the late 1980's, but per capita incomes would

also equalize. If the 1970-76 rates of change in per capita income in the South

and the non-South were to continue, the South would catch up with the

non-South in 13 years (in 1989) if no allowance is made for regional

cost-of-living differences. If the South's lower cost of living were taken into

account (which is not feasible), the catch-up would occur at an earlier date.

The purpose of these illustrations is not to predict when—or even

whether—the South will achieve parity with the rest of the Nation in poverty

levels and per capita income. The purpose is simply to underscore recent

regional income changes and to graphically describe their potential impact.

Many factors could, of course, interfere with this hypothetical timetable.

For one thing, future reductions of poverty in the South May be harder to

come by, for in the past the southern poor were more likely than the northern

poor to be workers whose wages were inadequate to provide more than a

poverty-level income. Many of these persons—sometimes called the “working

poor”—have benefited from rapid expansion in manufacturing and nonmanu

facturing employment in the South in the 1970's (for data, see tables 1 and 3

of Sternlieb and Hughes, 1977) and have taken jobs that enabled them to

become nonpoor. But as these processes run their course, a larger proportion of

the southern poor will consist of the nonworking poor—persons who are aged,

ill, handicapped, with dependent children at home, or otherwise not likely to

be available for employment. A relatively large part of the North's poor already

consists of persons belonging to these categories, and the South will find that

as the poverty population is reduced to its “hard core,” continued reduction in

numbers of the poor will be increasingly difficult to accomplish simply through

economic growth.

23

Another consideration is whether the South's shift to net inmigration of the

poor will put a damper on the rate of expansion of personal income. It could,

but the effect May not be too great. Recall that the South seems to have shifted

to net inmigration of the poor through reductions in outmigration rather than

attracting more of the poor from other regions. Throughout the 1967-77

period, there was a farily steady and small stream of low-income migrants to

the South, and this stream has not changed in magnitude. The South's recent

income gains have come at a time when the region was retaining more of its

own poor and converting them to nonpoor status. Eventually, the outmigration

stream of the southern poor Fell below the inmigration stream, giving the South

a small net inmigration of persons below the poverty level. An increased

inmigration of the nonsouthern poor would, of course, tend to depress future

increases in southern per capita income.

Summary and Conclusion

This paper developed data for use in examining the regional distribution and

Sources of redistribution of the Nation's poverty population. By disaggregating

annual interregional migration flows into poverty and nonpoverty components

(based on migrants’ income around the time of moving), the paper's most

important empirical result was to detect the southern region’s shift from

annual net outmigration to net inmigration of the poor between 1967 and

1977. Concomitantly, the Northeast, where many migrants from the South had

gone in previous decades, came to have a small annual net outmigration of

persons below the poverty level. The West and North Central regions appear to

be continuing patterns in existence inthe late 1960's, with the West having net

inmigration and the North Central region net outmigration of persons below

the poverty level.

One should bear in mind that there is no reason to suppose that every State

in the South and West has net inmigration of the poor or that every State in

the Northeast and North Central regions has net outmigration of the poor. The

various sample surveys that were used were not large enough to permit an

examination of the poor and nonpoor components of migration streams for

individual States.

An important result of the work reported here was to provide new findings

on the South’s growing volume of net inmigration. Although the southern

region as a whole shifted to net inmigration in the 1960's, the net inmigration

in that decade consisted entirely of the White nonpoor population. Not until

the 1970's has the South come to have net inmigration of Blacks and persons

below the poverty level. Clearly, the region's growing volume of net

inmigration in this decade is much more heterogeneous than before.

24

There is considerable irony in discovering that the South, still a region

where welfare benefits are generally quite low, has come to have net

inmigration of persons who are below the poverty level when they migrate.

These results certainly call into question theories about the poor being

attracted to areas where welfare benefits are high. Recent changes in migration

patterns of the Nation's poor population seem to derive largely from the

South's retaining more of its low-income population and converting them to

nonpoor status as a result of rapid growth in jobs. Presumably because of the

growth in jobs, the South has accounted for almost all of the Nation’s

reduction in the number of poor persons since 1970.

25

REFERENCES

Bowles, Gladys K., A. Lloyd Bacon, and P. Neal Ritchey. 1973. Poverty

Dimensions of Rural-to-Urban Migration: A Statistical Report. Washington,

D.C.: Economic Research Service, U.S. Department of Agriculture.

—, Calvin L. Beale, and Everett S. Lee. 1977. Net Migration of the

Population, 1960-70, by Age, Sex, and Color. Washington, D.C.: Economic

Research Service, U.S. Department of Agriculture.

, and James D. Tarver. 1965. Net Migration of the Population, 1950-60,

by Age, Sex, and Color. Volume II. Washington, D.C.: Economic Research

Service, U.S. Department of Agriculture.

Bowman, Mary Jean, and Robert G. Myers. 1967. “Schooling, Experience, and

Gains and Losses in Human Capital Through Migration.” Journal of the

American Statistical Association 62 (September): 875-898.

Bretzfelder, Robert B. 1977. “State Personal Income, 1975-76.” Survey of

Current Business 57 (August): 14-31.

Business Week. 1976. “The Second War Between the States.” May 17.

Business Week. 1977. “The $60 Billion Welfare Failure.” January 17.

Eldridge, Hope T., and Dorothy Swaine Thomas. 1964. Population Redistri

bution and Economic Growth, United States 1870-1950. Vol. III. Demo

graphic Analyses and Interrelations. Philadelphia: American Philosophical

Society.

Fein, Rashi. 1965. “Educational Patterns in Southern Migration.” Southern

Economic Journal 32 (July): 106-124.

Garnick, Daniel H. 1978. “A Reappraisal of the Outlook for Northern States

and Cities in the Context of U.S. Economic History.” Paper presented at the

Second Annual Conference on the Economic Future of the Northeast States,

Cambridge, Massachusetts, January 25.

Havemann, Joel, Neal R. Pierce, and Rochelle L. Stanfield. 1976. “Federal

Policies Cause North-South Shift.” National Journal 8 (June 26): 878-891.

27

Jusenius, C.L., and L.C. Ledebur. 1976. “A Myth in the Making: The Southern

Economic Challenge and Northern Economic Decline.” Economic Develop

ment Administration, U.S. Department of Commerce.

Long, Larry H. 1974. “Poverty Status and Receipt of Welfare among Migrants

and Nonmigrants in Large Cities.” American Sociological Review 39

(February): 45-56.

, and Lynne R. Heltman. 1975. “Migration and Income Differences

between Black and White Men in the North.” American Journal of Sociology

80 (May): 1391-1409.

_, and Kristin A. Hansen. 1975. “Trends in Return Migration to the South.”

Demography 12 (November): 601-614.

, and 1977a. “Interdivisional Primary, Return, and Repeat

Migration.” Review of Public Data Use 5 (March): 3-10.

, and . 1977b. “Selectivity of Black Return Migration to the South.”

Rural Sociology 42 (Fall): 317-331.

Lyons, Richard. 1976. “House Group Wants to Stem Sunbelt Flow.” The

Washington Post, September 2.

Moynihan, Daniel P. 1978. “The Politics and Economics of Regional Growth.”

The Public Interest 51 (Spring): 3-21.

National Center for Social Statistics. 1970. Public Assistance Statistics. April

1970. Washington, D.C.: Department of Health, Education, and Welfare.

—. 1976. Public Assistance Statistics. April 1976. Washington, D.C.:

Department of Health, Education, and Welfare.

Orshansky, Mollie. 1965. “Counting the Poor: Another Look at the Poverty

Profile.” Social Security Bulletin, January, pp. 3-30.

Sternlieb, George, and James W. Hughes. 1977. “New Regional and Metro

politan Realities of America.” Journal of the American Institute of Planners

43 (July): 227-241.

Time. 1975. “Unequal Welfare.” December 1, p. 8.

U.S. Bureau of Economic Analysis. 1973. Long Term Economic Growth

1860-1970. Washington, D.C.: U.S. Government Printing Office.

28

U.S. Bureau of the Census. 1973. Census of Population: 1970. Subject

Reports. Final Report PC(2)-9A, Low-Income Population. Washington, D.C.:

U.S. Government Printing Office.

. 1976a. Current Population Reports, Series P-20, No. 101. “Money

Income in 1974 of Families and Persons in the United States.” Washington,

D.C.: U.S. Government Printing Office.

. 1976b. Current Population Reports, Series P-25, No. 640. “Estimates of

the Population of States with Components of Change: 1970 to 1975.”

Washington, D.C.: U.S. U.S. Government Printing Office.

. 1977. Current Population Reports, Series P-60, No. 107. “Money

Income and Poverty Status of Families and Persons in the United States:

1976.” Washington, D.C.: U.S. Government Printing Office.

U.S. News and World Report. 1976. “‘Cities are Becoming Dumping Grounds

for Poor People’: Interview with Pierre de Vise.” April 5, pp. 55-65.

29

Superintendent of Documents

FIRST-CLASS MAILU.S. Government Printing Office

Washington, D.C. 20402

Official Business

Penalty for Private Use, $300

POSTAGE & FEES PAID

CENSUS

PERMIT NO. G–58